Saint-Saëns planar- AKA SMJ-40- time for a project now?

[technodanvan]

Looking really, really good Jon!

[technodanvan]

Meniscus had a pair of AudioPur PT2522C-4 that I grabbed purely because of the deep back cup. No idea what is packed in there (if anything) but I assume the tweeter is the same as the GRS. They appear to have a picture of a B/G on the site, so I'm curious if it'll be any different.

[JonMarsh]

I’m actually considering my MOD1 version for a rear firing tweeter on the MF-HF module. I’ve got a test for that almost ready… just to evaluate it again.

[JonMarsh]

you know what they say, you pays your money and you takes your chances!!

i’m doing phone support for my wife, straighten out some issues on her S23 Samsung (I am not an Android fan) and to make up for it I’m doing more milling work on the Saint-Saëns front panels! That will put me in a better mood, I think! Making some good progress, no yield loss either, so far!

[JonMarsh]

Speaking of tweeters, I'm building two "experiments" to evaluate for the rear firing tweeter- just variations in the rear sub enclosure. Let's call them "MOD5" and "MOD6". I'm also trying out a couple of different adhesives; so far, I still prefer the original big tube JBWeld or the smaller faster setting combined dual tube version. But the part should work...

The stock PT2522c has to have the glue on the back holding the rear cup slit and the cup pried off, then cleaned up. Also glue has to be removed from the body of the tweeter.







The cut pieces of phenolic sheet are glued to the back, and after setting up a bit, my favorite denim insulation is installed.




Next will be gluing on the backs, and letting them cure overnight.

I should have impedance sweeps and test results some time tomorrow...

[JonMarsh]

The LF module front panels are slow work, but now they're very close to where I want things. The driver rebate distance has been altered from 0.95" to 0.75", and the cutout routes have been finished, and the round over routing updated.

I need to update the CAD drawings for the rebate depth.






Next will be completing the pass through hole cutouts and beveling the sides.

[technodanvan]

What's the difference between MOD5 and MOD6? Deeper cup? More/less denim? Do you use anything to keep the back itself from resonating?

Edit: Apologies for all the questions...but at what point could we argue we should just use the open back variant and create a 6" deep subenclosure packed with denim?

[JonMarsh]

It all depends...

• MOD5 and MOD6 are different depths, as you surmised- which means creating a deeper sub enclosure of greater volume. Purpose is attempting cleaner absorption of back wave and less reflected to the front which is possibly a key issue in some of the presence range peaks and dips.
• The factory back cup is used in these experiments; it is curved and has damping material applied. I have also built ones with just a solid piece of phenolic, but they tend to create more of a plane wave reflection, it seems, based on measured response.
• The problem I see with the open back approach (not recommended by B&G, not used by PS Audio) is that the diaphragm is, perhaps you would say, somewhat "loose", perhaps not optimally tensioned. As a result, open back, the Fs is quite low (about 600Hz) and IB open back there is a 10 dB drop in level from the 12-16kHz area to say, 2500Hz. "Fixing" this, in comparison to the PT2522c, which is too small a rear volume, is what resulted in MOD1 through MOD4. Though I got the nominal SPL smoother between that and the crossover, I decided in the end to go with the TPL75. For a rear firing tweeter, my standards aren't as high, and the frequency range is different. It's an experiment, you know... results pending.
• The POC crossover is using a few tricks, including a series resistor to make the load impedance 8 ohms, and an 8 ohm Fostex Lpad to allow adjusting the output level. The higher Q impedance bumps of the IB part are not desired...

Does this remind you of anything? Maybe an AMT driver Curt used to use, whose production characteristics changed?


A refresher- this is how the MOD4 looks at 20 deg off axis- I'm just wondering if I can smooth out the area between 1 and 2 kHz a bit, by moving Fs lower, without raising the Q of the impedance peak significantly

Also, I'm aiming for a compound crossover, with a first order roll off at about 7500, steepening to 3rd order around 3kHz.

[technodanvan]

Interesting, thanks for the detailed answer Jon. I'm curious to see how the deep cup on the 'AudioPur' measures.

[JonMarsh]

I'd be curious to know that also, for reference comparison.

[JonMarsh]

Of course, as we know, slow work takes time, but eventually some of it gets done!







It's probably a bit boring watching "the sausage get ground", but it may give an idea of the amount of work to expect.


Next will be gluing the inner and outer front panels together... very precisely!

[technodanvan]

Looks like you recessed them a fair amount, any particular reason? Style?

The deep cups came in today, they do indeed say B|G right on them. The cup is plastic and about two inches deep. I haven't cut one off yet as I wanted to see if you would be interested in measuring one or both first. I can pop them in the mail this weekend if so. Parts Express does appear to sell the B|G version...at more than twice the price, of course.

[JonMarsh]

I can mount and measure them if you’re interested. I like the shape concept, would be hard to do DIY, of course. And you say PE has these too? Hmmm, I must not be spending enough time there.

The woofers are rebated sufficient distance to mount grille panels without fear of the driver surround or cone striking them. Originally it would have been less, with a 1-1/4” outer front panel, but with adding 1/4” maple panels to facilitate the cutouts for the tweeter to both top and bottom modules, well, they got a bit deeper.

[JonMarsh]

I’ll PM you my address tomorrow morning.. a skosh busy at the moment

[technodanvan]

No rush at all!

[JonMarsh]

If that’s what PE is selling as the deep back cup Neo3, all I can say is that the data sheet reference they have for it is somewhere between irrelevant and nonsensical and useless. No accurate physical description, a useless SPL plot alluding to a standard model, near illegible Xaxis numbers.

Color me curious, but concerned…

[technodanvan]

I didn't pull that up...very weird datasheet. It does seem to indicate the tweeter is 2.5 inches deep, which would be about right...but other than that, with no photos of the back, that is an odd one.

[JonMarsh]

I’d want to mount it on a GRS front plate, if you don’t mind. Have a bunch of them.

[technodanvan]

I’d want to mount it on a GRS front plate, if you don’t mind. Have a bunch of them.

Of course, I'd be using a GRS faceplate as well so no worries about scratching them or anything.

[JonMarsh]

You have a PM- I can donate two of the GRS front plates to your cause!

[Efalegalo]

I like the shape concept, would be hard to do DIY, of course.

But easy enough to 3D print those enclosures, and customize them to any volume needed (within reason).

[technodanvan]

Yet one more reason I should get a 3D printer, this would be a perfect project for it!

[JonMarsh]

Yet one more reason I should get a 3D printer, this would be a perfect project for it!

Yup! If I wasn't already awash in investments for which I'm working at realizing the results, I'd be thinking about that too.

[JonMarsh]

Between the heat, and so many other none audio tasks on the agenda, I have to keep strange hours to squeeze in a bit of work!

Both inner-outer front panel sets are now glued up (West system 105 epoxy) and alignment is to 1/32" tolerance-

Proprietary process involving machinist rules, and.... mechanical pencils!




Next will be basic sanding, then moving back to working on the upper module front panels. I even picked up a new remarkably low priced Bosch random orbital sander for some of this work- I swear, as many Bosch tools as I'm using on this project, I should work that into the project title somehow!

[technodanvan]

That looks pretty fantastic Jon. I wish I had heard of West Systems 12 hours ago, I just ordered some epoxy for the Keramiskas project and didn't have nearly as many options for fillers and such! Do you just use it straight?

[JonMarsh]

Hey Danny! I've been using this stuff for ages, it's just so flexible and convenient, depending on what you need to do- in my "recovery" operations on the CA storage unit I brought back a box of mixing cups for it I probably bought 10 years ago! The Isiris were put together with straight 105/205 combo, and lots of other projects. I am usually just using it straight, but the consistency is very flexible and useful for mixing with other stuff- not too runny, but easy to work with. The only thing requiring a bit of care is that it is exothermic, so you want to mix and apply reasonably quickly- any left standing in a mixing cup will harden up quicker than you expect, and faster than a spread layer in a joint bond.

This is my favorite "general purpose" epoxy, JBWeld Steel is my favorite "specialist" epoxy for unusual materials and strength requirements. You can tell it's different than most, they don't advertise this, but it's basically a marine epoxy, smells just like the marine epoxy I've used in the past, so it must be the same chemicals, I think.

[technodanvan]

Jon, due to the holiday there's a bit of a delay, but the package should be there Thursday.

[JonMarsh]

That sounds fine, Danny! With all the different "honey-do's" and other projects, Thursday sounds perfect! I'll setup and test these plus my recent builds the same day.

[technodanvan]

Hey Danny! I've been using this stuff for ages, it's just so flexible and convenient, depending on what you need to do- in my "recovery" operations on the CA storage unit I brought back a box of mixing cups for it I probably bought 10 years ago! The Isiris were put together with straight 105/205 combo, and lots of other projects. I am usually just using it straight, but the consistency is very flexible and useful for mixing with other stuff- not too runny, but easy to work with. The only thing requiring a bit of care is that it is exothermic, so you want to mix and apply reasonably quickly- any left standing in a mixing cup will harden up quicker than you expect, and faster than a spread layer in a joint bond.

This is my favorite "general purpose" epoxy, JBWeld Steel is my favorite "specialist" epoxy for unusual materials and strength requirements. You can tell it's different than most, they don't advertise this, but it's basically a marine epoxy, smells just like the marine epoxy I've used in the past, so it must be the same chemicals, I think.

So I had a second chance on what epoxy to use - mine came from Amazon and half the resin had leaked into the box. HUGE mess. So I got a refund on that and did a little research given my application, plus a little math to figure out how much I would really need.. I ended up getting MAS Epoxies "Deep Pour" Epoxy Resin. Since the pour I'm trying to do will be about 11" thick, that could cause problems with normal epoxy. As it is, it sounds like maybe I should accomplish it in several pours.

[JonMarsh]

Yeah, that sounds like it's right up the alley you need to travel! And the quantity, too!

One of my favorite features about the System West stuff is the metering pumps for the Resin and hardener- once you set it up, you get nice remeasured balanced amounts into one's mixing cup with just matching number of plunges for the resin and hardner- with the amounts I'm usually doing at one time, that's either 1 plunge or 2.

[JonMarsh]

Got some detailed sanding done early in the morning, but much of the day was all about putting together a Weber Spirit II BBQ grille that my past employer gave me in 2020 as an anniversary present. WAYYY more parts to assemble than the last one like that...

[technodanvan]

Yeah, that sounds like it's right up the alley you need to travel! And the quantity, too!

One of my favorite features about the System West stuff is the metering pumps for the Resin and hardener- once you set it up, you get nice remeasured balanced amounts into one's mixing cup with just matching number of plunges for the resin and hardner- with the amounts I'm usually doing at one time, that's either 1 plunge or 2.

Yeah, MAS seems to have that too...but they were out of the 3:1 pumps! Fortunately I don't think it will be necessary. I purchased four 1.3 quart mixes to ensure I would have enough; my plan is to simply mix the entire 1.3 quarts with sand, then pour about half of the mixture into each speaker and see where we're at. Might just stop there for the day, might mix and pour more.

The only issue with this whole thing is the heat. I can't really do it indoors due to the stink, so doing it in the garage is a must. Guess I'll just store everything indoors overnight, then pour in the morning and hope for the best. At least I don't need to worry about bubbles really.

Got some detailed sanding done early in the morning, but much of the day was all about putting together a Weber Spirit II BBQ grille that my past employer gave me in 2020 as an anniversary present. WAYYY more parts to assemble than the last one like that...

That'll be a good one at least, it should last you a long time.

[theSven]

Yeah, MAS seems to have that too...but they were out of the 3:1 pumps! Fortunately I don't think it will be necessary. I purchased four 1.3 quart mixes to ensure I would have enough; my plan is to simply mix the entire 1.3 quarts with sand, then pour about half of the mixture into each speaker and see where we're at. Might just stop there for the day, might mix and pour more.

The only issue with this whole thing is the heat. I can't really do it indoors due to the stink, so doing it in the garage is a must. Guess I'll just store everything indoors overnight, then pour in the morning and hope for the best. At least I don't need to worry about bubbles really.

Are you sure about the bubbles? I would think you should still blowtorch to be one the safe side and make sure all bubbles come out...

[JonMarsh]

How exothermic is that stuff? You'd want low exothermic long setting time epoxy...

[technodanvan]

It'll definitely still heat up. The guy pouring 9 gallons of it used a deep freeze to keep things cool - but it basically stopped the setting process. He needed to do single pours to eliminate seams between layers. Since I don't have that problem I'm thinking multiple pours is best - mix one batch, split it between the speakers, let it set for a day or two, repeat.

[technodanvan]

I suppose the bubbles could impact potential density, so maybe I should care a little bit. They're generally very small and most people get rid of them for cosmetic reasons though. I watched a video of a guy mixing and pouring nine GALLONS of this epoxy and he indicated that if I were to mix it and let it rest for 20 minutes before pouring most bubbles should go away. Then just pour slowly to minimize any new bubbles. I'm not sure I can wait a full 20 minutes when it's 90 degrees outside by 0800, but I can hit it with a blow dryer or something to speed things along.

I'm also not sure how the bubbles are affected by sand. This should be an 80/20 mixture (sand/epoxy, by weight). Basically, you get your 80% of sand, add the resin to it and mix. Then when you're ready, add the hardener and mix again. In general, it seems epoxy is around 1kg/l, so 1.3kg of epoxy to 10.4kg of sand in my case. I guess I need a scale of some kind...and I think I need more than the one 50 pound bag of all purpose concrete sand I already have.

What I'm getting from all this is doing it at scale - like, for a tower speaker - may be more cost prohibitive than Jon's methods. Sand is incredibly cheap ($6/50lbs), but the epoxy is fairly pricy. About $160 for what I think should be more enough for two 'small' bookshelf speakers. I think one would need approximately 8x this amount for a large set of towers. Granted, buying larger containers would be a bit more cost effective than the more manageable size I ordered.

[Steve Manning]

Danny you might want to check out this site. https://www.youtube.com/@BlackForestWoodCo They have developed their own deep pour epoxy. Either way, they have a ton of info on the process.

[technodanvan]

Thanks Steve, looks like these guys have a lot of experience!

[JonMarsh]

Danny's bargain Neo3 Deep Cups arrived, and they look pretty nice!



In front I placed one of the GRS mounting kits- makes life a lot easier. Will be putting that on both Neo3's today.


Now, it's interesting to just look at them side by side....

The Neo3 is on the left, the GRS2522 on the right. (this is a MOD5 version)





Why do I say that?

A few things I notice... Well, it's pretty clear the stampers for the metal work have been updated- the opening size is slightly smaller for the GRS and the vertical spacing as regards metal is a bit different.

But possibly a more significant difference is that the GRS seems to use the PDR construction technology, whereas the deep cup Neo3 does not.




I did measure the impedance curve and some parameters of the Neo3 Deep Cup and the MOD5 2522 update.


NEO3




2522 MOD5




Fs behavior is near identical, but it's interesting to note the differences in some calculated parameters. Also, pure speculation, I note that the GRS part has higher amplitude "disturbances" in the impedance curve above 2kHz, and it wouldn't surprise me if this correlates with response irregularities in comparison to the Neo3 Deep Cup.

So, I'm going to go slightly out on a limb, and speculate that the 2522 may show better off axis response than the Neo3 Deep Cup, but the Neo3 Deep Cup may have smoother response in the top end.

By Sunday I should know- Saturday is heavily booked with local events outside the home (Slavic Festival, early dinner at "Schnitzel Garten") and today was already pretty booked up- been up since 4 AM to try to keep up with my task agenda!

Some day I've got to get the hang of this retirement thing... seems like it might be a while yet!

[theSven]

Tom Petty had it right - You never slow down you never grow old!

[technodanvan]

Big thing I notice is the use of rivets versus screws to hold the thing together, I'm not sure I would have picked up on the difference in the grill.

If these are indeed Neo3 drivers, Meniscus still has a pair of regular back plate and open back tweeters for the same price I got these.

I'm looking forward to the results!

[JonMarsh]

Should be up some time Sunday- Saturday (today) has a lot of wifey events- Slavic Festival, early dinner at favorite German-Ukraine restaurant. Yeah, I noticed the rivets, too. Have to keep what Meniscus has on hand in mind- depending on results. You get first dibs, of course!

[JonMarsh]

The basic cut out steps were completed for the through holes for tweeter and midrange in the 1-1/2" thick panel; the template driven technique was key. That'e being extended to the cut out holes being the template for each 1/2" thick BB ply driver mounting plate...



These will be matched to specific front panel, though the actual tolerances are very tight for "hand work". (No CNC setup here, folks!).

A detail view of the lower left corner of one panel...






And with a GRS PT5010 front panel dropped into the cavity...




Pretty fair fit for manual work, if I do say so myself! (Ouch! I think I just sprained my shoulder from patting myself on the back!)



The bevel work necessitates a router table approach; the Bosch model I ordered is supposed to be here this Monday.

(maybe I can get a Bosch endorsement for all the Bosch tools I've used on this project? )

🤣

Now I've got to get to work on the back panel fabrication, and cutting the side panels and interior enclosure and damping pieces out, but that will likely not happen until Tuesday....

(uh oh! That reminds me of something from my childhood...)





J. Wellington Wimpy: "I'll gladly pay you Tuesday for a hamburger today!"

[JonMarsh]

Back to the PT2522 MOD5 vs B-G Neo3 Deep Cup...

I have them mounted with GRS front panel adapters... ready to mount for testing, but with my wife's Lexus electrical problems, I've been sidetracked.

This photo might be interesting... any guess which is the GRS and which is the B-G?




The GRS built parts all fit like the one on the left- like a glove. The B-G parts are a bit off.

For now, I decided to just mount one B-G part, and also send an extra FP kit back to Danny with the other one.

I just used the supplied gasket from the Neo- it is thinner and easier to work with, but I'm going to pass along two of the 2522 gaskets to Danny.

Now, the way I do this mounting is with the screws also, but a big dot of JB Weld in all 6 of the machined recesses of the backside of the front panel.


that's sort of committing one, and not a great back seal, either, but I don't plan to use these in spaces which are "pressurized" with other signals.

Testing delayed until tomorrow.

[technodanvan]

The insides behind the grills also look different. Is this the lack of the PDR tech the GRS has?

[JonMarsh]

Yes, as I described earlier. It's harder to see in photographs than in real life, but it's more apparent in the photos I took today. The grill holes are slightly different sized and vertical spacing.

[JonMarsh]

The LF and MF-HF cabinet designs have been getting more "specific", that is you might say, incorporating in the CAD drawings elements I've been mulling over- in this case, mainly the addition of "damper pads" on the inside of the cabinet to reduce possible wall resonance and reduce reflections in high pressure zones.

Ahem...

Well, that's the idea, though given that I don't want to stretch this development out too much longer, let's just say one concept I'm keeping in mind is, above all else, do no harm!


😏


Hopefully the end results will indicate whether these things are, to borrow the title from an old TV series I enjoyed that's been revived, "Justified".


This render of the LF cabinet with no outside walls should give some clarity on the basic idea, addressing the panel spaces right behind the woofers:



The most interior layer is high density 3/4" felt, the base layer includes BK2 phenolic plate and Second Skin damping material on top of that; the phenolic to be affixed to the side walls using Weicon 310 MS poly adhesive.


Here's an alternative view with the left wall rendered also.





This render is the Midrange-High Frequency module update, with a modified dimension set for the midrange sub enclosure, and damper panels added to the side walls and back wall. The side walls only use 1/4" high density felt; the rear uses 3/4".




Of course, it should be kept in mind that all of this is an "experiment", based on some smaller sets of experiments but also on inferences, deductions, and plain old fashion "guesstimates"!

😜

[technodanvan]

I think this approach is a lot easier - and probably nearly as effective - as the box within a box strategy, especially on a large floorstander. Which Second Skin product did you use, just the Damplifier/Damplifier Pro? I used this on my little hatchback over a decade ago and it worked well. I totally forgot about this company.

[JonMarsh]

It's the Damplifier Pro. I selected and tried this material first for the CC Supreme POC cabinet build.

But of course, I'm cheating a bit, as I altered the midrange sub enclosure design so that it doesn't touch/glue to the main cabinet side walls, so with small effort I do have an enclosure within an enclosure. Have to do that anyway to get the right rear volume size and shape (in my previous 4 experiments) for the PT5010. Now, the damping structures are just to hopefully provide better control for the enclosure wall and lower the potential interaction with the driver. Of course, it will be stuffed with the ATS Denim.

If you need some small pieces of the ATS Denim for the midrange enclosures, let me know- I can send that to you.

[technodanvan]

Just shot you a PM Jon.

[JonMarsh]

Today I finished putting together the new bench top router table, after getting some different hardware for completing the assembly (due to review info and video on Amazon), and then did the deep bevel routing as planned for the PT5010 midrange driver, to achieve the ideal Z axis location in consideration of it's amplitude and phase response for crossover summing, as discussed perviously.

I used both of my larger Whiteside 1/2" shank 45 degree bevel bits, in consideration of whether this would be a good idea to do the initial cutting with the smaller bit, then finish up the large area cutting with a quite sharp new bit. In retrospect, this seems to have been unnecessary, given the low feed ranges and wood removal that I used- one would have been fine.




The relatively high precision pass through hole cut in the from panel is used as the guide for the routing operation raising the bit in small steps to increase the depth of the cut while using the inside of the through hole as the guide.





Here is the result after several passes at increasing cutting depth, seen from the front side.




after the router bit and table insert are swapped out, the last several passes can be made, until only 1/2" of the vertical through hole remains.





Flipping the front panel pieces over, the finished midrange deep bevel meets expectations and conforms to the CAD design. Some light detail sanding is all that is needed to finish up this operation.

[JonMarsh]

You most likely don't realize it, but I'm not managing this whole "retirement" thing all that well, and it seems like there's always a bazillion things to do! Besides just speakers!

But sometimes friends and acquaintances step up to help with things, so we're still making a bit of progress today on the front panels with the help of Kaptain Kluge and his minion assistant (sometimes I think the minion does most of the work...)




Anyway, I showed him where the 1/2" BB ply reserve stock was, and he got to work... first assessing the outer sizing dimensions to fit in the front panel, then cutting out the pass through hole for the PT5010 and checking it.





Then, all that was left was cutting the outer dimensions of the plate that fit into the front panel "template" route opening, checking that, and gluing it in... I think they used cream colored JBWeld wood epoxy, which doesn't drip much and has a working time of about 15 minutes.


Voila!




Of course, only the best parchment paper was used to keep things clean and tidy! Minions are good at that!

[JonMarsh]

The proof in the pudding is whether the assembled PT5010 with faceplate mounts properly as regards clearances...


This is the first PT5010 I put together with FP mounting kit for early testing....





"Houston, we have clearance for assembly and liftoff!"

[theSven]

JonMarsh , The progress is looking amazing 😍. You've clearly proven retirement is boring and hobbies like this are way more fun to partake in 😉

[JonMarsh]

Well, retirement can be defined in a lot of different ways- I have a lot of saved up/deferred goals, so playing golf and watching sports on TV are not part of my retirement!


Much of today has been spent doing maintenance including a battery swap for my wife's Lexus. These modern cars need to have system voltage maintained while you pull the old battery out, so there is an added degree of complexity. Sure does make me love the simplicity of my FJ Cruiser.


The PT5010 mounting baffle was trimmed and installed in the second MF-HF module front panel assembly today...



And I was asked to clarify what the backside of this panel looks like after this assembly operation, so....





This is the panel assembled yesterday.

Next steps for these assemblies will be cleaning up the veneer work a bit, then cutting the side bevels.

[gbegland]

I'm loving all the planar action going on around here these days. Once you get past the scary looking native freq response of these drivers, there's no denying the actual sound quality once properly EQ'd and implemented is off the charts. There are just so many distortion mechanisms that aren't present in planar drivers compared to traditional moving coils drivers.

I'm working on getting a pair of the Radian LM10n drivers for testing in an ultra nearfield setup. Once that project is farther along, I'll start a separate thread.

Greg

[theSven]

Looking forward to your up coming thread on the Radian LM10n drivers.